From: Bhuvanchandra DV <bhuvanchandra.dv@toradex.com>
To: Richard Weinberger <richard@nod.at>, <linux-mtd@lists.infradead.org>
Cc: Artem Bityutskiy <dedekind1@gmail.com>
Subject: Re: UBI/UBIFS corruptions during random power-cuts
Date: Tue, 2 May 2017 10:17:43 +0530 [thread overview]
Message-ID: <dc7dce73-c83f-ad5b-a8de-bee951c6afbe@toradex.com> (raw)
In-Reply-To: <48d2f2b5-28e4-8d90-5fe6-6e3e4cbc383d@toradex.com>
Hello Richard,
On 02/16/2017 09:46 AM, Bhuvanchandra DV wrote:
> On 02/15/2017 08:44 PM, Richard Weinberger wrote:
>
>> Am 15.02.2017 um 16:06 schrieb Bhuvanchandra DV:
>>> On 02/15/2017 02:42 PM, Richard Weinberger wrote:
>>>
>>>> Am 15.02.2017 um 08:22 schrieb Bhuvanchandra DV:
>>>>> I was wrong, systemd version which we use lately is enabled with
>>>>> access control lists(ACL).
>>>>> ...
>>>>> [ 3.324846] systemd[1]: systemd 230 running in system mode.
>>>>> (+PAM -AUDIT -SELINUX +IMA -APPARMOR +SMACK +SYSVINIT +UTMP
>>>>> -LIBCRYPTSETUP -GCRYPT -GNUTLS +ACL +XZ -LZ4 -SECCOMP
>>>>> +BLKI)
>>>>> ...
>>>>>
>>>>> Will do the test with a systemd without ACL support.
>>>> Well, do files have xattrs?
>>>> See getfattr.
>>>> I'm trying to understand why your system makes so different.
>>> systemd's journald uses ACL, the journald's logs files seems to have
>>> extended attributes.
>>>
>>> root@colibri-vf:~# getfattr -R
>>> /var/log/journal/3e283b492ea046a29683c0801ceaba91
>>> getfattr: Removing leading '/' from absolute path names
>>> # file:
>>> var/log/journal/3e283b492ea046a29683c0801ceaba91/system@000547ec00621f43-a0f5cd8e864efc36.journal~
>>> user.crtime_usec
>>>
>>> # file:
>>> var/log/journal/3e283b492ea046a29683c0801ceaba91/system@000547ec25f38147-3ede62d61071df3c.journal~
>>> user.crtime_usec
>>>
>>> # file:
>>> var/log/journal/3e283b492ea046a29683c0801ceaba91/system@0820e4037f504446ad35bb362be7b809-0000000000000001-l
>>> user.crtime_usec
>>>
>>> # file: var/log/journal/3e283b492ea046a29683c0801ceaba91/system.journal
>>> user.crtime_usec
>>>
>>> # file:
>>> var/log/journal/3e283b492ea046a29683c0801ceaba91/system@37549ee306da42479500fb717cb2dada-0000000000000001-l
>>> user.crtime_usec
>>>
>>> # file:
>>> var/log/journal/3e283b492ea046a29683c0801ceaba91/system@000547ecafe0f654-e0c2574255c05aed.journal~
>>> user.crtime_usec
>>>
>>> # file:
>>> var/log/journal/3e283b492ea046a29683c0801ceaba91/system@000547ec78bce8a1-02b48f08c5d6a410.journal~
>>> user.crtime_usec
>>>
>>> # file:
>>> var/log/journal/3e283b492ea046a29683c0801ceaba91/system@0005488c6421da38-c4690dc0bb71068f.journal~
>>> user.crtime_usec
>> Hmm, not sure whether systemd journal triggers such a bug in UBIFS.
>> Anyway, can you provide me a UBIFS image after a failure?
>> A dd from the UBI volume (i.e. /dev/ubiX_Y) will do it.
>>
>> Maybe it can give me some hint.
>
> Here[1] is the raw image of the corrupted UBI volume.
>
>
> [1] https://share.toradex.com/xbczmjz0b2pntfv
The UBI/UBIFs corruption issue is still there even, with/without xattar,
we disabled systemd ACL support and ran the power-cut tests.
Tests are done on our down stream kernel and mainline kernel.
Here are the logs:
Kernel v4.4.59 without ACL:
https://pastebin.com/8MUrzqsL
Mainline kernel 4.11.0-rc4 without ACL:
https://pastebin.com/g0qJN7rJ
--
Bhuvan
>
> --
> Bhuvan
>
>>
>> Thanks,
>> //richard
>
prev parent reply other threads:[~2017-05-02 4:52 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-01-12 13:31 UBI/UBIFS corruptions during random power-cuts Bhuvanchandra DV
2017-01-12 14:12 ` Richard Weinberger
2017-01-12 14:14 ` Bhuvanchandra DV
2017-01-13 9:31 ` Bhuvanchandra DV
2017-01-13 11:15 ` Richard Weinberger
2017-01-13 13:26 ` Bhuvanchandra DV
2017-01-13 11:18 ` Richard Weinberger
2017-01-13 13:27 ` Bhuvanchandra DV
2017-01-18 5:11 ` Bhuvanchandra DV
2017-01-18 7:58 ` Richard Weinberger
2017-01-19 5:36 ` Bhuvanchandra DV
2017-01-19 9:08 ` Richard Weinberger
2017-01-25 6:29 ` Bhuvanchandra DV
2017-01-26 8:32 ` Richard Weinberger
2017-01-27 6:00 ` Bhuvanchandra DV
2017-01-30 5:48 ` Bhuvanchandra DV
2017-01-30 7:57 ` Richard Weinberger
2017-01-30 13:31 ` Bhuvanchandra DV
2017-01-30 20:48 ` Richard Weinberger
2017-01-31 7:06 ` Bhuvanchandra DV
2017-01-31 8:31 ` Richard Weinberger
2017-01-31 8:31 ` Bhuvanchandra DV
2017-01-31 8:38 ` Richard Weinberger
2017-02-06 6:10 ` Bhuvanchandra DV
2017-02-08 22:10 ` Richard Weinberger
2017-02-09 4:59 ` Bhuvanchandra DV
2017-02-10 13:01 ` Richard Weinberger
2017-02-13 3:34 ` Bhuvanchandra DV
2017-02-13 7:57 ` Richard Weinberger
2017-02-13 11:45 ` Bhuvanchandra DV
2017-02-15 7:22 ` Bhuvanchandra DV
2017-02-15 9:12 ` Richard Weinberger
2017-02-15 15:06 ` Bhuvanchandra DV
2017-02-15 15:14 ` Richard Weinberger
2017-02-16 4:16 ` Bhuvanchandra DV
2017-05-02 4:47 ` Bhuvanchandra DV [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=dc7dce73-c83f-ad5b-a8de-bee951c6afbe@toradex.com \
--to=bhuvanchandra.dv@toradex.com \
--cc=dedekind1@gmail.com \
--cc=linux-mtd@lists.infradead.org \
--cc=richard@nod.at \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox